Searcharxiv⌕ Search

arXiv subjects

Iain R. Moyles

Publications and source records attributed to Iain R. Moyles.

3 recordsLinked to original sources

An Asymptotic Analysis of Space Charge Layers in a Mathematical Model of a Solid Electrolyte

We review a model for a solid electrolyte derived under thermodynamics principles. We non-dimensionalise and scale the model to identify small parameters, where we identify a scaling that controls the width of the space-charge layer in the electrolyte. We present asymptotic analyses and numerical solutions for the one dimensional zero charge flux equilibrium problem. We introduce an auxiliary variable to remove singularities from the domain in order to facilitate robust numerical simulations. From the asymptotics we identify three distinct regions: the bulk, boundary layers, and intermediate layers. The boundary and intermediate layers form the space charge layer of the solid electrolyte, which we can further distinguish as strong and weak space-charge-layers respectively. The weak space-charge-layer is characterised by a length, $λ$, which is equivalent to the Debye length of a standard liquid electrolyte. The strong space-charge-layer is characterised by a scaled Debye length, which is larger than $λ$. We find that both layers exhibit distinct behaviour, we see quadratic behaviour in the strong space-charge-layer and exponential behaviour in the weak space-charge-layer. We find that matching between these two asymptotic regimes is not standard and we implement a pseudo-matching approach to facilitate the transition between the quadratic and exponential behaviours. We demonstrate excellent agreement between asymptotics and simulation.

physics.app-ph↗

Asymptotic reduction, solution, and homogenisation of a thermo-electrochemical model for a lithium-ion battery

We study two thermo-electrochemical models for lithium-ion batteries. The first is based on volume averaging the electrode microstructure whereas the second is based on the pseudo-two-dimensional (P2D) approach which treats the electrode as a collection of spherical particles. A scaling analysis is used to reduce the volume-averaged model and show that the electrochemical reactions are the dominant source of heat. Matched asymptotic expansions are used to compute solutions of the volume-averaged model for the cases of constant applied current, oscillating applied current, and constant cell potential. The asymptotic and numerical solutions of the volume-averaged model are in remarkable agreement with numerical solutions of the thermal P2D model for (dis)charge rates up to 2C, and reasonable agreement is found at 4C. Homogenisation is then used to derive a thermal model for a battery consisting of several connected lithium-ion cells. Despite accounting for the Arrhenius dependence of the reaction coefficients, we show that thermal runaway does not occur in the model. Instead, the cell potential is simply pushed closer to the open-circuit potential. We also show that in many cases, the homogenised battery model can be solved analytically, making it ideal for use in on-board thermal management systems.

physics.app-ph↗

Asymptotic reduction of a porous electrode model for lithium-ion batteries

We present a porous electrode model for lithium-ion batteries using Butler--Volmer reaction kinetics. We model lithium concentration in both the solid and fluid phase along with solid and liquid electric potential. Through asymptotic reduction, we show that the electric potentials are spatially homogeneous which decouples the problem into a series of time-dependent problems. These problems can be solved on three distinguished time scales, an early time scale where capacitance effects in the electrode dominate, a mid-range time scale where a spatial concentration gradient forms in the electrolyte, and a long-time scale where each of the electrodes saturate and deplete with lithium respectively. The solid-phase concentration profiles are linear functions of time and the electrolyte potential is everywhere zero, which allows the model to be reduced to a system of two uncoupled ordinary differential equations. Analytic and numerical results are compared with full numerical simulations and experimental discharge curves demonstrating excellent agreement.

physics.app-ph↗